Document 7772072

Download Report

Transcript Document 7772072

FLAT s atomnames
The atoms named in atomnames are restrained to lie on a
common plane within the standard uncertainty s (default
value 0.1 Å3).
CHIV V s atomnames
The chiral volumes of the named atoms are restraint to the
value of V within the standard uncertainty of s (default
value: 0.1 Å3. The default value for V is 0.
The chiral volume is defined as the volume of the
tetrahedron formed by the three bonds to an atom. The
sign of the chiral volume is determined by the alphabetical
order of the atoms forming the three bonds.
E.g. the chiral volume of the alpha carbon in an L-amino
acid residue is ca. 2.5 Å.
Restraints on Displacement Parameters
SIMU and DELU take into account that atoms, which are
bound to one another, move similarly, both in direction and
amount.
ISOR assumes approximate isotropic behavior for otherwise
anisotropically refined atoms.
Both SIMU and DELU are based on physically very sensible
assumptions and can be used on almost all atoms in a model
when the data-to-parameter-ratio is low or other problems with
the refinement make this desirable.
SIMU should not be applied uncritically to very small ions and
atoms that are part of freely rotation groups.
DELU s1 s2 atomnames
This rigid bond restraint is applied to all bonds connecting to
atoms mentioned in atomnames. It restrains the ADPs of
two atoms in the direction of the bond between them to be
equal within the standard uncertainty s1 (default 0.01).
If no atomnames are given, all atoms are understood.
SIMU s st dmax atomnames
Atoms closer to one another than dmax (default: 1.7 Å) are
restraint to have the same Uij components within the
standard uncertainty of s (default value: 0.04). For terminal
atoms st is assumed (default: 0.08). If no atomnames are
given, all atoms are assumed.
SIMU is much bolder an assumption than DELU (hence the
much larger standard uncertainty).
ISOR s st atomnames
The Uij values of the atoms mentioned in atomnames are
refined to behave apprximately isotropic within the standard
uncertainty s, or st for terminal atoms (default 0.1 and 0.2).
If no atomnames are given, all atoms are understood.
ISOR can be useful for solvent molecules, esp. water, for
which SIMU and DELU are ineffective.
DELU, SIMU, ISOR
Other Restraints
The SUMP command allows to linearly relate several free
variables:
SUMP c sigma c1 m1 c2 m2 ...
applies the following linear equation to the specified free
variables:
c = c1∙fv(m1) + c2∙fv(m2) + ...
where c is the target value for the restraint and sigma the
standard uncertainty. c1, c2, etc. are weighting factors and
frequently 1; m1, m2, etc. refer to the individual free variables.
Other Restraints
DEFS globally changes the default standard uncertainties for the
following restraints: CHIV, DANG, DELU, DFIX, FLAT, SADI, SAME
and SIMU using the following syntax:
DEFS sd[0.02] sf[0.1] su[0.01] ss[0.04] maxsof[1]
In parentheses are the default values.
Other Restraints
DEFS sd[0.02] sf[0.1] su[0.01] ss[0.04] maxsof[1]
sd is the default for s in DFIX and SADI, and for s1 in the SAME
instruction; for DANG twice the value of sd is applied. sf is the
default standard uncertainty for CHIV and FLAT, su is the default
value for s1 and s2 in DELU, and ss is the default value for s in
SIMU.
The default value for st in SIMU and ISOR, as well as s2 in SAME
are calculated from the respective s or s1 values (unless
specified differently by the user).
maxsof specifies the maximum value up to which a site
occupation factor is allowed to refine to. Fixed site occupation
factors and sofs linked to free variable are not restricted by
maxsof.
Free Variables
In general, any parameter P or any DFIX, DANG, or CHIV
restraint can be defined in the .ins file as
10 ∙ m + p
There are four different cases:
m = 0: the parameter P with the starting value p is refined freely.
m = 1: the value of p is fixed and not refined at all.
m > 1: P = p ∙ fv(m)
m <-1: P = p ∙ [fv(-m)-1]
where fv(m) is the value of the mth free variable.
Free Variables
10 ∙ m + p
m = 0: the parameter P with the starting value p is refined freely.
Trivial: describes a refinable parameter P as possessing the
starting value p.
Free Variables
10 ∙ m + p
m = 1: the value of p is fixed and not refined at all.
Assume you want to constrain an atom to lie on a mirror plane
parallel to the a-b plane at c = -¼. The task is to fix the value for
the z coordinate to -0.25.
According to the above, this can be done by giving m the value
of 1, and the value for p should be the atomic parameter of z
(i.e. -0.25).
Hence, the atomic parameter for z in the .ins file for this atom
reads 9.75.
Free Variables
10 ∙ m + p
m = 1: the value of p is fixed and not refined at all.
To give a second example: Sometimes it can be helpful to fix the
isotropic displacement parameter of an atom, U, at a certain
value, for example 0.05.
As always when parameters are fixed: m = 1; and p is the
desired value for U: 0.05.
The site occupation factor for the atom in question is then given
as 10.05.
Free Variables
10 ∙ m + p
m > 1: P = p ∙ fv(m)
m <-1: P = p ∙ [fv(-m)-1]
where fv(m) is the value of the mth free variable.
This involves additional free variables.
Most common case: disorder.
Linking the occupancy of an atom to the second free variable,
instead of the first one:
sof:
11.0000
11.0000


21.0000
-21.0000
Free Variables
10 ∙ m + p
m > 1: P = p ∙ fv(m)
m <-1: P = p ∙ [fv(-m)-1]
where fv(m) is the value of the mth free variable.
CHIV and the distance restraints DFIX and DANG can also be
combined with free variables:
Free Variables
CHIV and the distance restraints DFIX and DANG can also be
combined with free variables:
E.g. restraining a ClO4- ion to be tetrahedral: Assuming the
atoms in the ion are named Cl(1) and O(1) to O(4), the restraints
using SADI are as follows:
SADI Cl1 O1 Cl1 O2 Cl1 O3 Cl1 O4
SADI O1 O2 O1 O3 O1 O4 O2 O3 O2 O4 O3 O4
Free Variables
CHIV and the distance restraints DFIX and DANG can also be
combined with free variables:
E.g. restraining a ClO4- ion to be tetrahedral: Using DFIX and the
second free variable in the same scenario:
DFIX 21 Cl1 O1 Cl1 O2 Cl1 O3 Cl1 O4
DFIX 21.633 O1 O2 O1 O3 O1 O4 O2 O3 O2 O4 O3 O4
This corresponds to m = 2 for the second free variable and p is
1.0 in the first line and 1.633 in the second line (taking into
account that the 1,3-distances in a regular tetrahedron are 1.633
times as long as the 1,2-distances).
The value of the second free variable is refined freely and will
converge at the mean Cl-O-distance.
Free Variables
E.g. restraining a ClO4- ion to be tetrahedral: Using DFIX and the
second free variable in the same scenario:
DFIX 21 Cl1 O1 Cl1 O2 Cl1 O3 Cl1 O4
DFIX 21.633 O1 O2 O1 O3 O1 O4 O2 O3 O2 O4 O3 O4
The value of the second free variable is refined freely and will
converge at the mean Cl-O-distance.
The advantage of the second way is that the average Cl-O
distance will be calculated with a standard uncertainty (in
addition to the individual Cl-O distances with their standard
uncertainties). The disadvantage is that one additional least
squares parameter is to be refined (the second free variable).
Crystallographic Results
The final .res file contains the complete anisotropic model with
all hydrogen atoms, which can be used to generate attractive
figures for scientific publications (or grant proposals) and gain
several kinds of information about a molecule. The most obvious
are bond lengths and angles, but numerous other quantities can
be calculated from the atomic coordinates, such as torsion
angles or hydrogen bonds.
Bond Lengths and Angles
If the command BOND appears in the header of the .ins file,
SHELXL writes into the .lst file a table of all bond lengths and
angles in the connectivity table. BOND $H expands this table
to include all distances and angles involving hydrogen atoms
as well.
Torsion Angles
If the crystallographer or chemist wishes certain torsion angles
to appear in a separate table in the .lst file, each of these
torsion angles can be specified in a CONF command:
CONF atomnames
where atomnames defines a covalent chain of at least four
atoms. If no atom names are specified, SHELXL generates all
possible torsion angles.
Planes
MPLA na atomnames
SHELXL calculates a least-squares plane through the first na
atoms of the named atoms. The equation of this plane,
together with the deviations of all named atoms from the plane
and the angle to the previous least-squares plane (if present)
is written into the .lst file.
If na is not specified, the program fits the plane through all
named atoms
Hydrogen Bonds
If the command HTAB appears in the header of the .ins file,
SHELXL performs a search over all polar hydrogen atoms
present in the structure and examines hydrogen bonding.
The bonds listed in the .lst file are those for which the distance
between acceptor and hydrogen atom are smaller than the
radius of the acceptor atom plus 2.0 Å, and the angle between
the donor atom, the hydrogen and the acceptor atom is larger
than 110°.
Hydrogen Bonds
HTAB donor-atom acceptor-atom
SHELXL generates hydrogen bonds with standard
uncertainties and, in combination with ACTA, the appropriate
table in the .cif file.
EQIV can be used to specify a symmetry equivalent of the
acceptor atom.
RTAB
RTAB codename atomnames
The command RTAB allows to compile a variety of structural
quantities. Depending on how many atoms are specified in the
qualifier atomnames the following is calculated (and
tabulated):
•
•
•
•
chiral volumes (one atom specified)
distances (two atoms)
angles (three atoms)
torsion angles (four atoms specified)
codename must be specified and serves as an aid to identify
the tabulated quantity in the .lst or .cif file. It must begin with a
letter and cannot be longer than four characters.
ACTA
If the command ACTA appears in the header of an .ins file,
SHELXL generates a .cif file.
ACTA automatically sets the BOND, FMAP 2, PLAN and
LIST 4 instructions.
ACTA cannot be combined with other FMAP or LIST
commands.
Torsion angles defined by CONF and hydrogen bonds defined
by HTAB are also written into the .cif file.
The quantities defined by RTAB and MPLA commands do not
appear in the .cif file. They can be found in the .lst file only.
Next Meeting
Tuesday Mai 10, 2005,
11:30 a.m.
AMDUR room (here)
Topic: Reciprocal Net